BACKGROUND AND PURPOSE: MR cisternography has been used as the noninvasive screening tool of the cerebellopontine angle. The purpose of this study was to directly compare two currently dominant types of sequences for heavily T2-weighted MR cisternography. METHODS: Three-dimensional fast asymmetric spin-echo (3D-FASE) sequences, which are 3D half-Fourier rapid acquisition with relaxation enhancement and 3D constructive interference in the steady-state (3D-CISS) sequences, were compared on a clinical 1.5-T MR unit using the same scan times. In five healthy volunteers, the contrast-to-noise ratio (C/N) between CSF and the cerebellum was measured at three locations. Then, for qualitative analysis, the quality of the labyrinth was scored on the original source multiplanar reformatted images, the virtual endoscopic images, and the maximum intensity projection (MIP) images. In 20 consecutive patients with suspected cerebellopontine angle tumors, visualization of the tumors was evaluated using 3D contrast-enhanced spoiled gradient-echo imaging as the standard of reference. RESULTS: Both sequences showed comparable mean C/N values; however, in qualitative analysis, the scores for 3D-CISS on the source, virtual endoscopic, and MIP images were significantly lower than those on the images obtained with 3D-FASE, owing to more prominent flow and magnetic susceptibility artifacts on the 3D-CISS sequences. In all subjects, discontinuity of the semicircular canals was seen on the virtual endoscopic and MIP images obtained with 3D-CISS, owing to susceptibility artifacts, but not on those obtained with 3D-FASE. All 12 tumors were detected by both sequences, but 3D-CISS gave one false-positive result. CONCLUSION: 3D-FASE is considered the method of choice because artifacts are reduced and specificity is increased.
BACKGROUND AND PURPOSE: MR cisternography has been used as the noninvasive screening tool of the cerebellopontine angle. The purpose of this study was to directly compare two currently dominant types of sequences for heavily T2-weighted MR cisternography. METHODS: Three-dimensional fast asymmetric spin-echo (3D-FASE) sequences, which are 3D half-Fourier rapid acquisition with relaxation enhancement and 3D constructive interference in the steady-state (3D-CISS) sequences, were compared on a clinical 1.5-T MR unit using the same scan times. In five healthy volunteers, the contrast-to-noise ratio (C/N) between CSF and the cerebellum was measured at three locations. Then, for qualitative analysis, the quality of the labyrinth was scored on the original source multiplanar reformatted images, the virtual endoscopic images, and the maximum intensity projection (MIP) images. In 20 consecutive patients with suspected cerebellopontine angle tumors, visualization of the tumors was evaluated using 3D contrast-enhanced spoiled gradient-echo imaging as the standard of reference. RESULTS: Both sequences showed comparable mean C/N values; however, in qualitative analysis, the scores for 3D-CISS on the source, virtual endoscopic, and MIP images were significantly lower than those on the images obtained with 3D-FASE, owing to more prominent flow and magnetic susceptibility artifacts on the 3D-CISS sequences. In all subjects, discontinuity of the semicircular canals was seen on the virtual endoscopic and MIP images obtained with 3D-CISS, owing to susceptibility artifacts, but not on those obtained with 3D-FASE. All 12 tumors were detected by both sequences, but 3D-CISS gave one false-positive result. CONCLUSION: 3D-FASE is considered the method of choice because artifacts are reduced and specificity is increased.
Authors: Y Shigematsu; Y Korogi; T Hirai; T Okuda; I Ikushima; T Sugahara; L Liang; M Takahashi Journal: J Comput Assist Tomogr Date: 1999 Mar-Apr Impact factor: 1.826
Authors: S Naganawa; T Itoh; H Fukatsu; T Ishigaki; T Nakashima; Y Kassai; M Miyazaki; H Takai Journal: AJNR Am J Neuroradiol Date: 1998-04 Impact factor: 3.825
Authors: S Naganawa; K Yamakawa; H Fukatsu; T Ishigaki; T Nakashima; H Sugimoto; I Aoki; M Miyazaki; H Takai Journal: Eur Radiol Date: 1996 Impact factor: 5.315